and ensuring control stability [123].
Convenience sampling (Удобная выборка) – Using a dataset not gathered scientifically in order to run quick experiments. Later on, it’s essential to switch to a scientifically gathered dataset.
Convergence (Конвергенция) – Informally, often refers to a state reached during training in which training loss and validation loss change very little or not at all with each iteration after a certain number of iterations. In other words, a model reaches convergence when additional training on the current data will not improve the model. In deep learning, loss values sometimes stay constant or nearly so for many iterations before finally descending, temporarily producing a false sense of convergence. See also early stopping.
Convex function (Выпуклая функция) – is a function where the area above the graph of the function is a convex set. The prototype of a convex function is U-shaped. A strictly convex function has exactly one local minimum point. Classical U-shaped functions are strictly convex functions. However, some convex functions (such as straight lines) do not have a U-shape. Many common loss functions are convex: L2 loss, Log Loss, L1 regularization, L2 regularization. Many variants of gradient descent are guaranteed to find a point close to the minimum of a strictly convex function. Similarly, many variants of stochastic gradient descent have a high probability (though not a guarantee) of finding a point close to the minimum of a strictly convex function. The sum of two convex functions (e.g., L2 loss + L1 regularization) is a convex function. Deep models are never convex functions. Notably, algorithms designed for convex optimization tend to find reasonably good solutions in deep networks anyway, even if those solutions do not guarantee a global minimum.
Convex optimization (Выпуклая оптимизация) – The process of using mathematical techniques such as gradient descent to find the minimum of a convex function. A great deal of research in machine learning has focused on formulating various problems as convex optimization problems and in solving those problems more efficiently. For complete details, see Boyd and Vandenberghe, Convex Optimization [124].
Convex set (Выпуклое множество) – A subset of Euclidean space such that a line drawn between any two points in the subset remains completely within the subset. For instance, the following two shapes are convex sets.
Convolution (Свертка) — The process of filtering. A filter (or equivalently: a kernel or a template) is shifted over an input image. The pixels of the output image are the summed product of the values in the filter pixels and the corresponding values in the underlying image [125]
Convolutional filter (Сверточный фильтр) – One of the two actors in a convolutional operation. (The other actor is a slice of an input matrix.) A convolutional filter is a matrix having the same rank as the input matrix, but a smaller shape.
Convolutional layer (Сверточный слой) – A layer of a deep neural network in which a convolutional filter passes along an input matrix.
Convolutional neural network (CNN) (Распознавательная нейронная сеть) is a type of neural network that identifies and interprets images
Convolutional neural network (Сверточная нейронная сеть) – In deep learning, a convolutional neural network (CNN, or ConvNet) is a class of deep neural networks, most commonly applied to analyzing visual imagery. CNNs use a variation of multilayer perceptrons designed to require minimal preprocessing. They are also known as shift invariant or space invariant artificial neural networks (SIANN), based on their shared-weights architecture and translation invariance characteristics. Сonvolutional neural network is a class of artificial neural network most commonly used to analyze visual images. They are also known as Invariant or Spatial Invariant Artificial Neural Networks (SIANN) based on an architecture with a common weight of convolution kernels or filters that slide over input features and provide equivalent translation responses known as feature maps.
Convolutional operation (Сверточная операция) – The following two-step mathematical operation: Element-wise multiplication of the convolutional filter and a slice of an input matrix. (The slice of the input matrix has the same rank and size as the convolutional filter.) Summation of all the values in the resulting product matrix.
Corelet programming environment (Среда программирования Corelet) is a scalable environment that allows programmers to set the functional behavior of a neural network by adjusting its parameters and communication characteristics.
Corpus (Корпус) – A large dataset of written or spoken material that can be used to train a machine to perform linguistic tasks.
Correlation (Корреляция) is a statistical relationship between two or more random variables.
Correlation analysis (Корреляционный анализ) is a statistical data processing method that measures the strength of the relationship between two or more variables. Thus, it determines whether there is a connection between the phenomena and how strong the connection between these phenomena is.
Cost (Стоимость) – synonym for loss. A measure of how far a model’s predictions are from its label. Or, to put it more pessimistically, a measure of how bad a model is. To determine this value, the model must define a loss function. For example, linear regression models typically use the standard error for the loss function, while logistic regression models use the log loss.
Co-training (Совместное обучение)
Co-training essentially amplifies independent signals into a stronger signal. For instance, consider a classification model that categorizes individual used cars as either Good or Bad. One set of predictive features might focus on aggregate characteristics such as the year, make, and model of the car; another set of predictive features might focus on the previous owner’s driving record and the car’s maintenance history. The seminal paper on co-training is Combining Labeled and Unlabeled Data with Co-Training by Blum and Mitchell [126].
Counterfactual fairness (Контрфактическая справедливость) – A fairness metric that checks whether a classifier produces the same result for one individual as it does for another individual who is identical to the first, except with respect to one or more sensitive attributes. Evaluating a classifier for counterfactual fairness is one method for surfacing potential sources of bias in a model. See “When Worlds Collide: Integrating Different Counterfactual Assumptions in Fairness” for a more detailed discussion of counterfactual fairness.
Coverage bias (Систематическая предвзятость охвата) this bias means that the study sample is not representative and that the data set in the array has zero chance of being included in the sample.
Crash blossom (Двусмыссленная фраза) – A sentence or phrase with an ambiguous meaning. Crash blossoms present a significant problem in natural language understanding. For example, the headline Red Tape Holds Up Skyscraper is a crash blossom because an NLU model could interpret the headline literally or figuratively.
Critic